ON _THE CONDITION WHICH DETERMINES THE CHEMICAL SIMILARITY O F E L E M E N T S AND RADICLES BY GEOFFREY M A R T I N
We propose to show in this paper that chemically similar elements attract the same radicles with pyoporlional intensities of force, whereas chemically unlike elements attract them with unlike intensities. Mathematically this may be expressed thus : If we denote the attractions which an element A exerts on the various radicles A, B, C, D . . . . by],, . . . .fn ; and the corresponding attractions which an element B exerts by f I, f,, fa, . . . . f n . Then, if the elements -4 and B are chemically similar, the following relation holds : (1)
Further we will show that the closer K approaches unity, the closer the elements approach each other in their chemical properties. When K = I the forces the two elements exert are equal and this is obviously the condition that the elements A and B may be perfectly cheinically similar, for in this case one may replace the other in any chemical reaction without altering the nature and direction of the reaction which is determined by these forces. When K = n the one element possesses the same chemical properties as the other, but intensified n-fold. For example for the system fluorine and chlorine, K has the value 1.7 as the following thermal data shows (see later):
(F, Si‘d) = 67.3 ==I,
(c1’ s1.:)
=39.4-ffl’
(F, H )
(Cl, H)
=22.0=f2’
-
where
= 37.6 = I 2
Szmilavzty of Elemmts aizd Radicles
563
This means that fluorine possesses the same chemical properties as chlorine, but intensified 1.7-fold. I n general, the relationship (I) is not satisfied by the affinities of any two elements A and B selected at random. We find, however, that the more closely their affinities conform to this law, the closer is their chemical resemblance. And conversely, the more widely their affinities depart from this law, the remoter is their chemical resemblance, so that in the case of elements which behave in a quite different way chemically, this relation ship is not even remotely satisfied, as we will presently show. X o two elements are perfectly chemically similar. All exhibit certain chemical peculiarities which distinguish one from the other. Therefore the relation (I) is never exactly satisfied by the affinities of two elements, but is the limiting condition which is more and more exactly fulfilled as the elements approach more and more exactly to perfect chemical similarity. For atoms of the same element alone is the condition exactly fulfilled, K here taking the value unity since the forces are now of the same intensity. It may be noted that many of the so-called radicles play the part of elementary atoms, although quite incapable of existing -N3, -CN), simply because separately (for example NH,-, they are arrangements of atoms so combined together as to exert forces of the same intensity as the elementary atoms, of which they play the part. The best way to make the above law evident, is to contrast the quantities of heat evolved when equivalent quantities of like and unlike elements combine with the same radicles. Of course, it must be remembered that the heat evolved in the formation of a compound, does not exactly measure the force driving the atoms together, except perhaps, at the absolute zero of temperature. Further, the heats of formation of compounds as given in
564
Geofrey iVavtiiz
the text-books are not the tvue heats of formation, but are merely dzference effects which may be positive or negative-where the true heats of formation must be always positive. For example, when hydrogen combines with iodine, the change is not simply
H+I=HI but is a double decomposition ; thus :
*I H
H +
!I 1)
=
I
+
Now, if the heat required to effect the separation of H from H, and I from I, is greater than the amount of heat generated when 2H combines with 21, then the result will be negative. Hence it is apparent that even in those cases of chemical action in which an evolution of heat is actually observed, this heat represents only the surplus energy remaining over after the breaking Lip of the molecules of the original substances employed in the experiment. I n fact it is only when the heat evolved in the formation of a compound is very great, that it can be taken as measuring approximately the attractive forces, for in this case the incidental errors become small in comparison with the quantity of heat evolved. Bearing these limitations in mind, the conformance to the above law which appears (as we will immediately show) when we contrast the quantities I have evolved when the atoms of chemically similar elements unite with the same radicle, must be regarded as nothing else than surprising. 1. Chemically Unlike Elements
When we contrast chemically unlike elements the relations
are not even remotely satisfied, and K varies widely and irregularly for every different ratio. As typical pairs of unlike elements, we will contrast H with C1, on the one hand, 0 with I on the other. In neither case are the relations satisfied.
Similavify of Elemeafs a?zd Radicles
545
Hydrogen and Chlorine ( H , Li) = 21.6 = f,’ (H, Na) = 13 = fy’ (H, C1/J = 5.4 = f,’ (H, Si’/,) = 8.2 = f,’ ( H , N1/8)= 4.0 = f,‘ 3.86 = f,’ ( H , P’/,) = ( H , AS’/,*)= - 1 2 . 2 == f,‘ (H, Sb’/,) = -28.1 = f,’ ( H , O’/>) 34.2 == f,’ (H, S‘/&)= 2.3 = flo’
Whence
L-
f,’ - --I
24 2.2
--
1.9
Hence K takes in succession each of the following values : 7.5, 2.3, 4 7 , -3.2, 5.5, -1.9,-I, -0.26, 2.6. 4-3, T h e ratio is in no case constant but varies widely and irregularly. T h e varying ,of K would be very much greater if we only knew the H.F. of all the corresponding compounds ; but as a matter of fact experimental difficulties limit the determination of the H.F. of compounds only to those which are stable enough to exist at an ordinary temperature. Many compounds of chlorine are known of great stability, whereas the corresponding hydrides are too unstable to exist at ordinary temperatures, e.g., PbC1, and PbH,.
L
( H , I) = -
(I, C u ) = (I, Ax) = (I, Li) = (I, Mgl/,) = (I, N a ) = '
(I. Hgl/,) = (1, K ) = ( I , Cal/,) -(I, Srl/J = (I, Cl) = (1, s>= ( I , As'/,) -= (I, Si1/,) = (I, Pb'/,) = Whence
K, therefore, takes any of the followiiig values : -5.7, 1.06, 1.65,0.66,0.99, 0.59,1.35,0.89,-4.3, 3.54,2 . 7 , 3.78,1.14, and hence is not even remotely constant. An examination of the corresponding compounds of any two dissimilar elements will reveal the same peculiarity. 0.21, 1.10,
Similavz’tj~of Elements and Radicles
567
2. Chemically Similar Elements U’hen, however, we contrast chemically similar elements, a very different relationship is observed. K remains approximately constant. Chlorine and Bromine (c1, CLl) = 32.9 = /, (Br, Cu) = 25 = f,’ (Cl, Ag) = 29.2 = f, (Br, Ag) = 2 2 . 7 = f,’ (Cl, Li) = 93.8 = f, .(Br, Zi) = 80 = f3’ (Cl, Na) =. 97.7 == f, (Br. Na) = 85.8 = f4’ (Cl, K) = 105.6 = f6 (Br, K) = 95.3 = f,’ (Cl, Call,) = 84.6 = j’, (Br, Cal/,) = 70.4 = f,’ (CI, Sr1j2) == 92.3 -=f i (Br, Srl/,) = 78.8 = f,’ (Cl, Bali,) = (Cl, Zn’j,) = (Cl, Hgl/,) = (Hg, C1) = (Cl, A1113)= (Cl, As1/,) = (C1, Sbl/,‘) = (C1, Pbl/,) =
Whence
97.3 =-: ), 48.6 = j 8 27.2
= f,,
3 2 . 6 == fll 53.66 = f,, 23.8 = f,, 30.46 = I,, 41.0 -=f I 5
(Br, Bal/,) (Br, Znl/,) (Br, Hgl/,) (Br, Hg) (Br, All/,) (Br, As1/,) (Br, Sbl/,) (Br, Pb’i,)
-=8 5 . 0 = 40.0 = 20.3 = 24.5 = 41.0 = 19.7
= Is’ = f,’ = ilo’ == f l l f = flz’ -= f13/
= 25.63 = f,,’ = 36.0 = lfl6’
568
Geofrey Martiit
K is here throughout the series approximately constant, never becoming greater than 1.34nor less than 1.11. When me contrast these values of K with the previous values, we see how closely constant K remains. T h e values of K range in the case of H and C1 from 7.5 to -3.2 and in the case of 0 and I, from -5.7 to 3.78.
B I -
:
16
:
:
:
:
K12 K13 K14 K15
-2 -
-3 -
Fig.
I
Showing the approximate cotistancy of K for similar elements (C1 and Br) and its variation for dissimilar elements (C1 and H ) . The erratic line CD records the successive values of K for the system C1 and H. The nearly straight line AB records the successive values of K for the system C1 and Br.
T h e approximate constancy of K for similar elements, and its variation for dissimilar elements is best demonstrated graphically.
”- . . . . . . K n for f7 n ; and ‘ I,’ ’ fn . . . . . K, as abcissae, and the succes-
Writing K, for IL * K, for
*
fl’
taking K,, K,, K,, K, sive numerical values as ordinates, we find for the system C1 and H an erratic line, for the system C1 and Br an almost straight line parallel to the axis of x (Fig. I).
Fluorine and Chlorine T h e only data available is the following: (F, Sil/J = 67.3 = f , (F, J3 = 37.6 = f2
(Cl, Si’/4)= 39.4 = f,’ (Cl, H ) = 2 2 = j 2 ’
Sinzilavity of Elemeizts am?RadicZes
569
Whence :
T h e ratio is, therefore, constant even for such different elements as H and Si. The large value of K shows, that F possesses the same chemical properties as C1, but intensified I. 70 times. Sodium and Potassium The only data available is the following: ( K , Cl) (K, Br) ( K , I) ( K , O’W (K, NH,)
= 105.6 = f , = 95.3 = = 80.1 = f d
I,
= 47.4 = f, = 6.3 = f ,
(Na, C1) ( N a , Br) ( N a , I) (Na, O’/J ( N a , NH,)
I,’ I,$’ = 45.5 = f,’ = 5 . 2 = Ij’ = 97.7 = = 85.8 = 12’ = 69.1 =
Whence : I3
I,’
- 80. I --
69.1
= 1.16
I t will be seen, that here also K is constant and very nearly equal to unity-a fact, which explains the great chemical resemblance of Na and K. Potassium a n d Caesium The only available data is the following: (Cs, Br) = 99.8 = f, ( K , Br) = 95.3 = f,’ (K, I) -- 80.1 = f2’ (CS, I) = 86.6 = iZ Whence ; L - 9 9 = A’ - 95.3 h.- 86.6 ~
f,’ - 80. I
I.oj = 1.08
Here again, K is constant and very nearly equal to unity. I n the case of all the alkali metals this is so, and this is the reason why there exists such a close resemblance between them.
Zinc and Cadmium The only data available is the following: (Cd, Cl,) = 93.0 = f,’ (Cd, 0) = 75.5 = j,’
(Zn, C1,) = 97.2 = f , (Zn, 0) = 85.4 = f, Whence :
Here again K is constant and nearly equal to unity, which accounts for the close resemblance of Zn and Cd. Boron a n d Silicon The case is of great interest, inasmuch as B and Si possess a great degree of chemical similarity, and yet B is trivalent and Si tetravalent. T h e chemical similarity of elements does not, in fact, depend upon their valence but solely upon their attractive forces. T h e following data-the only available-shows, that K is nearly equal to unity, a fact which explains the remarkable chemical resemblance of these two elements. (Si1/,,C1) = 39.4 = f , O1/J = 54.8 = f ,
(B1/5, C1) = 34.70 = f,’ (B1/8, 01/,)= 52.8 = f,’ Hence :
f,’
= 39.4
34.7
-
1.13
iz = 54.8 = 1.04 f2’
52.8
Caleium and Strontium (Cal/,, CI) = 84.6 = fl’ (Srl/,, C1) = 92.3 = j , (Cali,, Br) = 70.4 = I,’ (,%I/,, Br) = 78.8 = f, (Cali,, I) = 53.6 = f,’ (SiA],, I) = 71.7 = f, (Cal/,, Ol/,) = 65.5 = f4’ (Srl/,, 01/,) = 64.2 = f, 1/4(Ca,0,) = 37.6 = f6’ 0’) = 35.5 = f,
Siwzilarily of Elements and Radzcles
571
Whence ;
k-7r.7-- 1.33
I,' f4-
f,'
- 53.6 -
64.2 = 0.98 65.5
~
Here again K is nearly equal to unity, a fact which explains the chemical resemblance of Ca and Sr. Strontium a n d Barium (Sr'j2, Cl) = 92.3 = fl' (Sr1/2,Br) = 78.8 = f,' (Sr1/2q I) = 71.7 = f,' (,%I/%, 01/*) = 64.2 = fl I/4(Sr, 0,) = 35.5 = f 5 '
(Ba'i,, Cl) = 97.3 = f, (Ball,, Br) = 8j.o = f, (Ba'/,> I) = 72.3 = f , (Ba'/,, O'/?)= 62.1 = f , '/,(Ba, 0,) = 34.0 = f b
Whence :
Here again K is nearly equal to unity, which accounts for the great chemical similarity of Sr and Ba.
Nickel and Cobalt T h e only available data is the following : C1) = 37.2 = fl' (Nil/,, Se1i2) = : 9.21= f2' Tel/,>= 7.55 = f3'
(Coli,, C1) = 38.2 = fl (Co'i,, Se'i,) = 9.64 = A (Co'/,, Tel/,) = 7.65 = j ,
Similariiy of Elements and Radicles
572 Whence :
Here again K is constant and very neat-j equz,, to unity, as we should expected from the remarkable chemical resemblance of Co to Ni. Antimony and Bismuth T h e data available is meagre : (Sb'/,, C1) = 30.46 = f, (SbZ,Os, 3H,O) = 167.52 = i,
(Bi1/3, C1) = 30.21 = j,' (Biz, O", 3H,O) = 137.74 = i2'
Whence :
T h e preceding pages contain practically all the available thermal data for comparing the corresponding compounds of chemically similar elements. I n every case where elements possess a great degree of similarity, K is approximately constant, and very often nearly unity. And in every case where elements differ widely in their chemical properties, precisely the reserje holds true. University of Kiel.
